Individual Multimodal Pathway Statistics for Predicting Treatment Response in Late-life Depression

点击翻译按钮获取中文摘要

Modest response rates to first-line antidepressant treatment for late-life depression (LLD) expose individuals to prolonged depressive symptoms that worsen their prognosis and associated health risks. Biomarkers of treatment response can alleviate this burden by identifying individuals most likely to benefit from antidepressant treatment. MRI measures of brain structure and function are a promising tool to identify such biomarkers, though the performance required for clinical translation has remained elusive. The goal of this proposal is to integrate complementary network measures from structural and functional MRI with clinical measures to generate biologically relevant features that can improve prediction of treatment outcome in LLD. The anticipated impact of this research will provide improved personalization of LLD treatment (NIMH Strategic Objective 3.2), while elucidating the neural circuitry indicative of treatment outcome (Objective 1.3). To achieve this goal, structural, resting state, and diffusion-weighted MRI will be collected from 75 participants with LLD before commencing an algorithmic antidepressant treatment protocol. The role of resting state functional connectivity as a mediator of the relationship between structural connectivity and clinical measures (baseline depression severity and change in depression severity over treatment) will be investigated within key neural circuitry at the group level. Individual Multimodal Pathway Statistics (IMPathS) will be derived to quantify the personalized importance of functional connectivity to the relationship between structural connectivity and depression severity for prediction of treatment outcome at the individual level. Utility of IMPathS will be assessed by their ability to improve performance beyond unimodal MRI and clinical predictors. Dr. Gerlach has a PhD in nuclear engineering and radiological sciences and is completing a transition from computational physics to computational neuroscience. He will require additional training in 1) the neurobiology, clinical manifestations, and treatment of LLD, 2) diffusion-weighted imaging processing and analysis, 3) advanced statistical training for development and testing of IMPathS, 4) human subjects, study design, and data collection. Completion of the training and research plan in this career development award will enable Dr. Gerlach to progress to an independent investigator focused on investigating the neurobiology of late life anxiety and mood disorders through improved integration of multimodal neuroimaging measures. Dr. Gerlach will execute this training and research with the full support of the Department of Psychiatry at the University of Pittsburgh, which is a highly collaborative environment focused on the development of early career scientists.